Classification on AICCE 1.0 (test) benchmark leaderboard · SOTA2 Research

Classification on AICCE 1.0 (test)

99.8Accuracyo1

Evaluation Results

Method	Links
o12026.04		99.8	100	99	99.5	21.45	4,050	—
o12026.04		96.7	99.7	96.1	97.8	0.338	—	3
gpt-4o2026.04		96.6	95.9	100	97.9	0.11	—	3
llama-3.3 70B2026.04		96.4	95.6	100	97.7	0.101	—	12
o12026.04		95.7	96.4	98.2	97.3	20.21	0	—
claude 4.5 sonnet2026.04		95.7	98.8	95.7	97.2	19.829	4,319	—
claude 4.5 sonnet2026.04		95.3	93.7	82	87.4	0.047	—	3
gpt-4o2026.04		94.1	95.6	97	96.3	5.147	1,776	—
Random Forest2026.04		93.8	85.57	83	84.26	0.0009	—	—
MLP Neural Net2026.04		93	80.95	85	82.93	0.0023	—	—
Logistic Regression2026.04		93	94.52	69	79.77	0.0004	—	—
llama-3.3 70B2026.04		93	92.2	99.7	95.8	6.466	0	—
Extra Trees2026.04		92.8	83.33	80	81.63	0.0004	—	—
SVM2026.04		91.4	98.31	58	72.96	0.0003	—	—
qwen-3-3b2026.04		91.2	99.3	89.6	94.2	10.81	0	—
gpt-4o2026.04		89.4	88.3	99.8	93.7	4.92	0	—
KNN2026.04		88.8	95.83	46	62.16	0	—	—
Naive Bayes2026.04		88.2	100	41	58.16	0	—	—
Gradient Boosting2026.04		85	60.33	73	66.06	0.0015	—	—
BERT + LSTM2026.04		84	100	20	33.33	0.3653	—	—
Isolation Forest2026.04		77.4	42.7	38	40.21	0.0007	—	—
XGBoost2026.04		56	29.87	89	44.72	0.0002	—	—
LightGBM2026.04		50	26.56	85	40.48	0.0002	—	—